use leveledb to store,but save failed

[busyfree]

• Ego version (or commit ref):
• Go version:go version go1.11 linux/amd64
• Gcc version: 4.9.4 (GCC)
• Operating system and bit:CentOS release 6.8 (Final) 64bit
• Provide example code:
  beego main.go init server

func main() {
	if beego.BConfig.RunMode == "dev" {
		beego.BConfig.WebConfig.DirectoryIndex = true
		beego.BConfig.WebConfig.StaticDir["/swagger"] = "swagger"
	}
	initSearcher()
	beego.BeeApp.Server.RegisterOnShutdown(
		func() {
			initClose()
		})
	beego.Run()
}

func initSearcher() {
	// 初始化
	var dictFile = filepath.Join(beego.AppPath, "conf/dictionary.txt")
	var stopTokenFile = filepath.Join(beego.AppPath, "conf/stop_tokens.txt")
	controllers.Searcher.Init(types.EngineOpts{
		Using:         1,
		GseDict:       dictFile,
		StopTokenFile: stopTokenFile,
		IndexerOpts: &types.IndexerOpts{
			IndexType: types.LocsIndex,
		},
		UseStore:    true,
		StoreOnly:   true,
		StoreFolder: "/tmp/shareyou_search",
		StoreEngine: "ldb",
	})

}
func initClose() {
	controllers.Searcher.Flush()
	controllers.Searcher.Close()
}

api post to save data into engine
api.go

package controllers
import (
	"fmt"
	"github.com/go-ego/riot/types"
	"lib/utils"
	"reflect"
	"strings"
)

var (
	Searcher         = riot.Engine{}
)

const (
	// SecondsInADay seconds in a day
	SecondsInADay = 86400
	// MaxTokenProximity max token proximity
	MaxTokenProximity = 2
)

type ApiController struct {
	BaseController
}

type DocItem struct {
	Id        string `json:"id"`
	Timestamp uint64 `json:"timestamp"`
	UserName  string `json:"username"`
	LikeCount uint64 `json:"count"`
	Text      string `json:"text"`
	DocId     string `json:"docid"`
}

// WeiboScoringFields  weibo scoring fields
type DocItemScoringFields struct {
	Id        int64
	Type      string
	Timestamp int64
	LikeCount int64
}

// WeiboScoringCriteria custom weibo scoring criteria
type WeiboScoringCriteria struct {
}

// Score score and sort
func (criteria WeiboScoringCriteria) Score(doc types.IndexedDoc, fields interface{}) []float32 {
	if reflect.TypeOf(fields) != reflect.TypeOf(DocItemScoringFields{}) {
		return []float32{}
	}
	wsf := fields.(DocItemScoringFields)
	output := make([]float32, 3)
	if doc.TokenProximity > MaxTokenProximity {
		output[0] = 1.0 / float32(doc.TokenProximity)
	} else {
		output[0] = 1.0
	}
	output[1] = float32(wsf.Timestamp / (SecondsInADay * 3))
	output[2] = float32(doc.BM25 * (1 + float32(wsf.LikeCount)/10000))
	return output
}


func (o *ApiController) Post() {

	content := o.Input().Get("content")
	typeStr := o.Input().Get("type")
	ts := utils.MsTimestampNow()
	id, _ := o.GetInt64("id", 0)
	count, _ := o.GetInt64("count", 1)
	sfId, err := SnowFlakeServer.NextID()
	docId := fmt.Sprintf("%d", sfId)
	Searcher.Index(docId, types.DocData{
		Content: content,
		Fields: DocItemScoringFields{
			Id:        id,
			Type:      typeStr,
			Timestamp: ts,
			LikeCount: count,
		},
	})
	Searcher.Flush()
	return
}

func (o *ApiController) Get() {
	keyword := o.Input().Get("keyword")
	output := Searcher.SearchDoc(types.SearchReq{
		Text: keyword,
		RankOpts: &types.RankOpts{
			ScoringCriteria: &WeiboScoringCriteria{},
			OutputOffset:    0,
			MaxOutputs:      100,
		},
	})

	docs := make([]*DocItem, 0, 0)
	for _, doc := range output.Docs {
		item := DocItem{}
		field, _ := doc.Fields.(DocItemScoringFields)
		item.DocId = doc.DocId
		item.Text = doc.Content
		item.Timestamp = uint64(field.Timestamp)
		for _, t := range output.Tokens {
			item.Text = strings.Replace(item.Text, t, "<font color=red>"+t+"</font>", -1)
		}
		docs = append(docs, &item)
	}
	return
}

Description

when ctrl+c to exit the server ,then run again, search the keyword ,but nothing find. any help?

[busyfree]

why leveldb opened at rito/engine.go ,but when write data into leveldb then return leveldb status is closed.

[busyfree]

Add a retry label to control the error.